In the female mammal, one of two X-chromosomes is inactivated to ensure that males and females have equal X-chromosome dosage. X-chromosome inactivation (XCI) can take place randomly or in an imprinted manner on the paternal X (XP). In the mouse, random XCI is observed in the embryo proper (soma) and imprinted XCI in the placenta. Both somatic and placental XCI require the 'X-inactivation center'(Xic), an X-linked region known for genes making noncoding RNA (ncRNA). The Xic produces Xist RNA, a large untranslated RNA that 'paints'the X and recruits silencing factors to the X in cis. Our laboratory previously identified the antisense Tsix gene and a second noncoding element, Xite, as essential repressors of Xist. In the previous funding cycle, we focused on the molecular means by which Tsix and Xite regulate different steps of XCI. This investigation has resulted in three major discoveries: (i) In females, the Xs 'pair'transiently with each other at the onset of XCI. We believe that pairing mediates 'cross-talk'between the Xs to ensure mutually exclusive chromosome fates. Tsix and Xite mediate the pairing interaction. (ii) Tsix controls Xist expression through preemptive changes in chromatin structure driven by its antisense transcription. Tsix RNA also binds Dnmt3a and directs de novo DNA methylation to the Xist promoter. (iii) Dosage compensation occurs much earlier in development than previously thought. We believe that imprinted XCI has already taken place by the time of conception and, on this basis, propose that the XP is inherited as a 'pre-inactivated'chromosome from the paternal germline. These discoveries have raised many new questions, which we now intend to pursue in three distinct Aims. In Aim 1, we will define molecular factors required for X-chromosome pairing. In Aim 2, we intend to investigate how chromatin modifications are directed by Tsix transcription and what chromatin modifiers are recruited by Tsix RNA. In Aim 3, we will carry out genetic and biochemical tests to distinguish between the pre-inactivation and de novo models for imprinted XCI. Information gained from the research program will have broad implications for human genetics, cancer, and epigenetic reprogramming. It will also enhance our understanding of basic biological processes, especially those involving chromosome dynamics, ncRNA regulation, and transgenerational inheritance.